Customer evidence
Completed assessment, approved financial and operating metrics, management priorities, and clarifying responses.
Methodology v2.1
A benchmark is an evidence system—not a customer survey.
The WealthOps Intelligence Engine™ combines multiple source layers, preserves uncertainty, and limits comparisons to cohorts the available evidence can support.
Multi-source evidence stack
Customer inputs are validated and contextualized—not accepted as the whole benchmark.
Company evidence
Company websites, service mix, locations, leadership, hiring activity, licenses, safety records, and other attributable public signals.
WealthOps peer evidence
Normalized, permissioned company observations accumulated through completed engagements and stored without exposing client-confidential information.
Industry context
Public datasets such as Census County Business Patterns, BLS QCEW, IRS industry statistics, EIA CBECS, OSHA, and applicable state licensing records.
Licensed evidence
Commercial data is used only when access, terms, field definitions, and permitted analytical use have been verified.
Benchmark controls
Every comparison carries its own audit trail.
Define the metric
Use a documented definition, unit, period, calculation rule, and allowed source types.
Normalize the evidence
Align time periods, company scope, geography, and units before comparison.
Select the cohort
Match relevant factors such as commercial HVAC focus, size band, geography, service mix, and period.
Test sufficiency
Record sample size, coverage, freshness, and comparability. Weak cohorts produce directional context—not a false percentile.
Version the benchmark
Store the cohort definition, source snapshot, effective date, and methodology version used in the result.
Review the conclusion
Check source support, scoring consistency, decision relevance, confidentiality, and unsupported precision.
Evidence language
The label must match what is actually known.
Provided by an authorized company representative; not independently verified unless stated.
Directly supported by an attributable external or internal source.
Calculated from documented inputs and assumptions.
A reasoned interpretation that is not directly confirmed.
Insufficient evidence; retained as a gap rather than filled with false precision.
Minimum result record
No score without provenance.
Source name, URL or internal record, and observation date
Metric definition, value, period, and evidence label
Cohort definition, sample size, and benchmark effective date
Methodology version and calculation rule
Confidence rating, limitations, and reviewer
Current-state disclosure
The methodology and data model are production standards. Benchmark depth depends on the evidence and cohort available for each metric. When a statistically defensible peer comparison is not yet available, the deliverable uses a documented operating reference, directional context, or “insufficient evidence”—never an invented percentile.
Public sources provide context and do not, by themselves, establish a company-specific conclusion. Licensed sources are not represented as active until contractual and permitted-use requirements are confirmed.
See the standard in use